Skip to main content
LanceDB is a multimodal lakehouse for AI teams that need one data layer for curation, feature engineering, search and retrieval, and model training. It is built on top of Lance, an open-source lakehouse format designed for multimodal AI data. Move from data exploration to model training on one unified platform without needing to manage a fragmented stack of storage, feature, retrieval, and training systems.

​
Train on the right data faster

Search, curate, and train at petabyte scale, with all your multimodal data in one table. Preparing training data and running experiments slow down when raw data, metadata, embeddings, features, and governance artifacts live in separate systems. LanceDB keeps them together in one versioned multimodal table, so AI teams spend less time stitching infrastructure together and more time improving datasets, testing features, and keeping GPUs fed.

​
Accelerate the data flywheel for model research

Keep your AI data in one table. Find the failure, fix the data, and retrain.
1

Find a curated training subset for every failure case

Start from one failed grasp or evaluation example. Search multimodal data, metadata, and embeddings together, retrieve similar examples, and curate them into a targeted training subset.
2

A single dataset for every modality

Keep episodes, frames, metadata, features, and embeddings together in one schema-enforced table, without pipelines to keep them in sync. Table versions let you trace each training run back to the exact data it used.
3

More hillclimbing, less babysitting preprocessing jobs

Define the computed columns and materialized views your experiment needs, such as extracted features or embeddings, and refresh them as your data changes. On LanceDB Enterprise, column refreshes run as background jobs on the cluster instead of in your code.
4

Higher GPU utilization, faster convergence

Keep GPUs fed. StreamingDataset streams multimodal training data into PyTorch with globally shuffled, deterministic batches, combining fast random access with high-throughput scans.
Use the same table to curate training data, add derived features, retrieve examples, and feed training jobs that rely on expensive GPUs. Training workloads can sample, shuffle, and scan projected columns from local storage or object storage, then assemble GPU-ready batches from a tagged dataset version. Training pipelines need curation, derived features, reproducible splits, fast random access, and a clean path into frameworks such as PyTorch — LanceDB keeps all of it connected through the same table model, whether a workflow uses one table or several related ones.

​
Architecture: Embedded and Enterprise

LanceDB ships as two offerings on the same Lance format and table model, which stores table data, versions, and indexes in open, columnar files. What changes between them is how storage and compute are organized around that format: start locally with Embedded, then move to Enterprise when your team needs distributed scale, managed infrastructure, private deployment, or higher-throughput curation, feature engineering, search and retrieval, and training workflows.

LanceDB Embedded

An open-source, embedded retrieval library that runs inside your application process, with no separate database server to deploy.

LanceDB Enterprise

A distributed multimodal lakehouse service for production workloads that need managed infrastructure and higher-throughput workflows.
See Embedded vs Enterprise for a feature-by-feature comparison, including architecture, latency, throughput, and day-to-day usage differences.

​
Projects using LanceDB

​
Training

  • stable-worldmodel — a platform for reproducible world-model research built on a LanceDB data layer, reporting 3-4x faster data loading on Push-T versus HDF5 / MP4 at a fraction of the disk footprint.
  • le-wm — a joint-embedding predictive world model from pixels, trained on the stable-worldmodel platform and its LanceDB data layer.
  • lerobot-lancedb — a drop-in LanceDB backend for HuggingFace LeRobot datasets with faster loading across robotics datasets.

​
Agentic applications

  • OpenClaw LanceDB Plugin — stores an agent’s long-term memory in LanceDB, automatically recalling relevant memories before a model turn and capturing new facts afterward.
  • Cognee — an open-source AI memory platform that gives agents persistent memory across sessions, using LanceDB as one of its vector store backends.
  • CrewAI — a multi-agent framework that rebuilt its agent memory on LanceDB.
Across these projects, LanceDB and Lance provide the multimodal data layer that keeps raw observations, annotations, features, and training or memory access patterns in one format instead of scattering them across task-specific stores.

​
Start with your workload

Ingest your data

Create tables, append records, and update or delete rows as your data changes.

Build search and retrieval

Use vector search, full-text search, hybrid search, reranking, filtering, and SQL.

Curate training data

Profile, sample, chunk, join, and deduplicate records before training on them.

Engineer features in place

Derive computed columns and materialized views without moving data elsewhere.

Load data into PyTorch

Use LanceDB tables and permutations for projected, shuffled, random-access training reads.

Manage catalog objects

Create, list, and drop tables, namespaces, and views from Python, TypeScript, or SQL.